I’ve struggled with anxiety for most of my adult life, and one of the very few tools I reach for on purpose, rather than by accident, is a long exhale. Breathe in, let the breath out for longer than feels natural, and wait for my shoulders to remember they’re allowed to come down.

In my head I had always filed this under “calming down.” And calming down, in the private logic of an anxious person, means becoming more careful. Smaller. Less likely to do anything rash.

So a study from Potsdam and Berlin, published in Neuron this May, felt a little like discovering that the brake pedal in my car was also wired to the accelerator (not that I’d have noticed, since I mostly use it while sitting very still). The same long exhale made people more willing to take a gamble.

Two seconds in, eight seconds out, and a coin toss on the screen

The team, led by Wenhao Huang and Soyoung Q Park at the German Institute of Human Nutrition Potsdam-Rehbruecke, worked with colleagues from Charité, Freie Universität Berlin and the German Naval Institute of Maritime Medicine. The journal article is paywalled, so I read their preprint, methods section and all. The design is almost disarmingly simple.

People lay in an fMRI scanner and decided, again and again, whether to accept a gamble: a coin toss that could win them between 10 and 30 euros or lose them between 5 and 15. Between blocks, only their breathing changed. In one condition they breathed at their own natural rhythm, worked out from a baseline recording. In the other, they followed a bar on the screen that filled as they inhaled and emptied as they exhaled, at what the preprint calls “a fixed ratio of 2:8 seconds.” That’s about six breaths a minute.

The team also tracked heart rhythm, skin conductance and pupil size, which makes this one of the more thorough attempts I’ve seen to watch a body and a decision in the same moment.

“Rather, our judgment emerges from the interplay between cognitive processes and our current bodily state.” — Soyoung Q Park

Forty-nine went in, and one fell asleep

The details I liked best only show up in the methods. Forty-nine people signed up. Eight were dropped: four couldn’t keep to the breathing rhythm in the scanner, one repeatedly fell asleep during the task, and three left for technical or medical reasons. As reviews of slow breathing go, falling asleep is hard to argue with. That left 41 people for the choices and 35 for the brain scans.

The money was real, but not much of it. Nobody found out during the task whether they’d won. At the end, three choices were picked at random and paid out at 10 percent, so the most anyone could win on one of them was 3 euros.

And because rewards ran from 10 to 30 euros while losses ran from 5 to 15, the game was tilted in the player’s favour. I counted: of the 121 pairings those ranges allow, 109 offered a bigger win than loss. Roughly nine in ten gambles were good bets on paper. The authors chose the 2:1 range on purpose, “to yield balanced choice distributions,” so people would say yes about as often as no. Still, it changes how I read the headline. Taking “more risks” here mostly meant accepting more offers that were worth accepting, which may be why the authors write that the shift “does not compromise decision quality.”

What moved was the pull of the win

Under the long exhale, people accepted more of the gambles. Two standard markers of the parasympathetic “rest” branch of the nervous system rose, both measures of how much the heartbeat varies from breath to breath and beat to beat. That’s exactly what you’d expect from slow breathing, and exactly why I use it.

That part isn’t new: a 2022 review of 223 studies found slow breathing raises this kind of heart rate variability fairly reliably.

The interesting part is how the extra risk-taking happened. If slow breathing simply made people less afraid, you’d expect it to show up in loss aversion, the well-documented tendency for a loss to sting more than an equal gain pleases. That’s the usual suspect whenever someone becomes bolder. But loss aversion didn’t differ between the two breathing patterns. Losing hurt just as much. Skin conductance and pupil size, two classic signs of arousal, didn’t budge either. What changed was the other side of the ledger: under the long exhale, the size of the possible win counted for more in people’s choices.

Huang put it plainly in the institutes’ announcement: “The interplay between breathing and cardiac dynamics makes the brain more receptive to rewards.”

The brain part is subtler than the headlines

One headline, on SciTechDaily, said slow breathing “can rewire your brain.”

Nothing was rewired, though. The scans measured activity during a few five-minute blocks of breathing.

I’d put the brain result more carefully too. In the preprint, it comes from comparing people with each other. Those whose heart rhythm shifted most under the long exhale showed the strongest reward-related activity in the ventromedial prefrontal cortex, a region closely involved in weighing how much something is worth, and in the precuneus. The preprint doesn’t report a group-wide jump in those areas. And when the team looked for a direct link between how much someone’s body changed and how much their choices changed, they found nothing strong or consistent.

So the chain from breath to heart to brain to choice is a plausible story with good pieces, not yet one unbroken line.

The accountant who never got quieter

I’ve been trying to work out why this finding unsettles me in a way I mostly enjoy. I think it’s because it quietly corrects a model of anxiety I’d been carrying around without ever examining it.

When anxiety is loud, it feels to me like an overcommitted accountant who only keeps one column: everything that could go wrong, itemized, cross-referenced, recalculated at 2 a.m. I’d assumed a long exhale worked by getting that accountant to put the pen down. This study suggests something different, at least for healthy adults making lab gambles. The loss column didn’t shrink. The gain column simply started to look bigger. The breath left the worry where it was and made room beside it for the possibility of something good.

Breathing and the anxious brain seem to talk in both directions. In 2024, a mouse study in Nature Neuroscience found a pathway from the front of the brain to the brainstem that slows breathing and calms anxiety-like behaviour at the same time. The Berlin study runs the arrow the other way, from a breath you choose to a heart that changes rhythm to a decision.

Jack Feldman, a UCLA neurobiologist who has spent decades studying the brainstem circuits that set our breathing rhythm, explains how breathing and emotion are wired together.

It also fits something from my own research on measuring affect, where much of the work comes down to treating bodily signals as data rather than as noise around the “real” psychological event. Here the body is one of the inputs to the decision, sitting right alongside the numbers on the screen.

How far I’d take it

Forty-one healthy, fairly young people, average age just under 25, is a small sample. Anyone with a history of psychiatric disorders was excluded, so nobody in this study was anxious by design, and my accountant reading is my own. The gambles paid out in single euros, in one session, in a scanner.

Nobody knows yet how long the effect lasts once you walk out and start breathing like a normal distracted person again, and the authors say future work needs to test it “across diverse populations and real-world situations.”

So I’d use a smaller word than the one Park reached for when she called the study “scientific proof” that breathing is “a reliable and targeted method capable of controlling our decisions.” Nudging is closer to what the data show.

Still, the design is solid: the same people chose under both breathing patterns, so the breathing itself is the most plausible cause of the shift. That’s more than most breath-and-mood headlines can claim, and still no reason to exhale slowly before your next big risk.

The next time I let the breath run long

I’ll keep using the long exhale. It still does the thing I need it to do, which is to get my body out of the emergency it has decided we’re in. But I think I’ll notice it differently now. I’d always treated it as a way of making the world feel less dangerous. It might be doing something slightly more generous: letting the good possibilities back into view at the moment the anxious part of me has stopped believing in them.

It’s a strange comfort, finding out that the eight seconds I spend trying to feel safer might also be the eight seconds in which I’m most willing to hope.

Note: This article is for general interest and does not constitute medical or psychological advice. Anyone whose anxiety is affecting their daily life is encouraged to speak with a qualified professional.

Sources: